News for hackers. Stuff that matters.
Re:Science fails to face the shortcomings of statistics by lutorm on Wednesday March 17, 2010 @05:32PM (#1199714)
"If I understand the CLT (big if) then the distribution of the _mean_ of a sample is, by the CLT, going to be Gaussian"

Yes, for an infinite number of samples. The rate at which is converges to a Gaussian, though, is strongly dependent on the distribution from which the measurements are drawn.

Re:Science fails to face the shortcomings of statistics by joe_the_user on Wednesday March 17, 2010 @08:40PM (#1200137)
Actually, not even this is not true!

The Central Limit Theorem applies only to samples which have finite mean and variance (see: http://en.wikipedia.org/wiki/Central_limit_theorem).

Take a distribution which has infinite variance or mean and you can wind-up instead with one of the fractal distributions which Mandlebrot studied.

see http://en.wikipedia.org/wiki/Stable_distribution

Distributions with infinite variance or mean are more common than one might imagine. Some might argue the stock market would qualify.

Re:Science fails to face the shortcomings of statistics by lutorm on Thursday March 18, 2010 @11:29AM (#1201523)
Good point. Distributions with infinite variance are less common in physics, so that's normally not something we have to worry about.